The Architectural Implications of Autonomous Driving: Constraints and Acceleration

By Shi-Chieh Lin, Yunqi Zhang, Chang-Hong Hsu, Matt Skach, Md E. Haque, Lingjia Tang, Jason Mars. 2018

488 citations recorded by the source corpus.

Autonomous driving systems have attracted a significant amount of interest recently, and many industry leaders, such as Google, Uber, Tesla, and Mobileye, have invested a large amount of capital and engineering power on developing such systems. Building autonomous driving systems is particularly challenging due to stringent performance requirements in terms of both making the safe operational decisions and finishing processing at real-time. Despite the recent advancements in technology, such systems are still largely under experimentation and architecting end-to-end autonomous driving systems remains an open research question. To investigate this question, we first present and formalize the design constraints for building an autonomous driving system in terms of performance, predictability, storage, thermal and power. We then build an end-to-end autonomous driving system using state-of-the-art award-winning algorithms to understand the design trade-offs for building such systems. In our real-system characterization, we identify three computational bottlenecks, which conventional multicore CPUs are incapable of processing under the identified design constraints. To meet these constraints, we accelerate these algorithms using three accelerator platforms including GPUs, FPGAs, and ASICs, which can reduce the tail latency of the system by 169x, 10x, and 93x respectively. With accelerator-based designs, we are able to build an end-to-end autonomous driving system that meets all the design constraints, and explore the trade-offs among performance, power and the higher accuracy enabled by higher resolution cameras.

Subfields: Aerospace Engineering, Automotive Engineering, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

In Automotive Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

Wireless Network Intelligence at the Edge (2018, neutral): “decisions for safe operation [12], [13].”
Milestones in Autonomous Driving and Intelligent Vehicles: Survey of Surveys (2023, unknown): “prediction [73], [74], planning [75], [76], [77], [78], [79], [80], E2E [81], [82], control [83], [84], [85], [86], [87], [88], [89], system [90], [91], [92], [93], hardware [94], [95], [96], soft-”
Edge-Computing-Enabled Smart Cities: A Comprehensive Survey (2019, unknown): “For example, more computing resources must be given to a self-driving car than AR to enable faster execution of the self-driving car functions to avoid accidents [164].”
Mobile Edge Intelligence and Computing for the Internet of Vehicles (2019, background): “Meanwhile, to achieve better safety than the best human driver, who takes action within 100∼150 ms, autonomous driving systems should be able to process real-time traffic conditions within a latency of 100 ms [23], which demands significant amount of computing power.”
Conditional Affordance Learning for Driving in Urban Environments (2018, background): “This is well within real-time requirements for autonomous driving systems, where latency should reach levels below 100 ms [30].”
MnnFast: A Fast and Scalable System Architecture for Memory-Augmented Neural Networks (2019, background): “Variants of existing DNNs usually have different requirements including new operations, which can motivate us to develop new optimized DNN accelerators [11, 15, 22, 39, 43, 64, 83].”
Edge Computing for Autonomous Driving: Opportunities and Challenges (2019, background): “[37] explored the architectural constraints and acceleration of the autonomous driving system.”
Eyeriss v2: A Flexible Accelerator for Emerging Deep Neural Networks on Mobile Devices (2018, background): “, 2017) ▪ Filter rows reused horizontally ▪ Ifmaps reused diagonally ▪ Psums accumulated vertically filter(1,2,3) x input(1,2,3) filter(1,2,3) x input(2,3,4) filter (1,2,3) x input (3,4,5) ofmap column 1 (1, row_size)”
AI-Native Network Slicing for 6G Networks (2021, methodology): “We consider a delay-sensitive autonomous driving service, e.g., object detection, whose delay requirement is 100 ms for road safety [14].”